prompt-evaluation-runn…
Use when evaluating prompts, LLM outputs, red-team suites, or model behavior with local eval configs and safe provider/cost controls.
Use when preparing release notes or a changelog entry from commits, PRs, or diffs, summarizing a sprint or version range, or scanning a range for breaking changes before publishing.
$ npx -y skills add yeaight7/agent-powerups --skill changelog-generator --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/changelog-generatorContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when preparing release notes or a changelog entry from commits, PRs, or diffs, summarizing a sprint or version range, or scanning a range for breaking changes before publishing.
name: changelog-generator description: Use when preparing release notes or a changelog entry from commits, PRs, or diffs, summarizing a sprint or version range, or scanning a range for breaking changes before publishing.
git log <from>..<to> --oneline --no-merges git log <from>..<to> --format="%h %s (%an)" --no-merges
Or from GitHub PRs:
gh pr list --state merged --base main --json number,title,labels,author
Check CHANGELOG.md for format precedent before writing.
| Commit signal | User-facing label | |---|---| | `feat` / new capability | Added | | `fix` / bug fix | Fixed | | `perf` / performance | Improved | | `deprecate` | Deprecated | | `BREAKING CHANGE:` footer or `!` suffix | **Breaking** | | `chore`, `refactor`, `ci`, `build` | (internal — omit from user notes) | | `docs` | Include only if docs are a shipped artifact |
When Conventional Commits are not used, infer from the commit subject.
Flag any commit or PR that:
**User-facing release notes** (GitHub Release body):
## What's Changed ### Added - Description of new capability ([#123](link)) ### Fixed - Description of bug fix ([#124](link)) ### Breaking Changes - `--old-flag` has been renamed to `--new-flag` — update your scripts ### Deprecated - `legacyMethod()` is deprecated; use `newMethod()` instead
**Internal changelog entry** (CHANGELOG.md format):
## [X.Y.Z] - YYYY-MM-DD ### Added ### Fixed ### Changed ### Breaking ### Deprecated ### Chores
**Breaking-change scan only:**
BREAKING CHANGES FOUND: abc1234 feat!: rename --output-dir to --out def5678 fix: remove deprecated --legacy flag NONE if clean.
**Technical diff summary** (for internal reviewers):
CHANGED FILES: N MODULES AFFECTED: [list] SUMMARY: [what changed at the file/module level, no prose]
Deliver the formatted block plus:
Inspired by `ComposioHQ/awesome-codex-skills` `changelog-generator/SKILL.md`. Rewritten in Agent Powerups style.
Curated power-ups for coding agents: skills, slash commands, MCP configs, hooks, AGENTS.md templates, and workflows for serious software engineering. Claude Code, Codex, Antigravity CLI, Cursor and more
Repo: yeaight7/agent-powerups
Use when evaluating prompts, LLM outputs, red-team suites, or model behavior with local eval configs and safe provider/cost controls.
Use when creating or reviewing red-team eval plugins, attack templates, grader rubrics, safety fixtures, or model-risk test metadata.
Use when designing, running, debugging, or hardening deterministic eval suites for agent skills, prompts, tool workflows, or MCP-backed cases.
Use when designing tool definitions for a new agent or subagent, an agent shows high retry rates, ambiguous tool invocations, or silent failures, or an…
Use when routing a prompt to a local provider CLI for a second opinion, review, or plan -- you are about to call a provider directly, need the response saved…
Use when starting work in an unfamiliar area of a codebase, spawning a subagent that needs targeted file context, a first search pass missed the relevant file,…